Extent of aortopulmonary collateral blood flow as a risk factor for Fontan operations

Between November 1987 and January 1990, 33 patients (tricuspid atresia, 9 patients; mitral atresia, 3; single ventricle, 15; others, 6) underwent Fontan operations. The rate of blood flow returning to the heart during aortic cross-clamping was measured as an indication of the extent of development of aortopulmonary collateral arteries. Percent cardiac return (calculated by dividing the blood flow rate returning to the heart by the cardiopulmonary bypass blood flow rate and expressing the value as a percentage), were 1% to 9%, 7 patients; 10% to 19%, 11; 20% to 29%, 9; 30% to 39%, 4; 40% to 49%, 1; and 50% to 59%, 1 patient. Percent cardiac return showed a significant correlation with postoperative mean systemic venous pressure ( r = 0.6, p < 0.01). In those patients in whom percent cardiac return was more than 33%, the mean systemic venous pressure after operation was high (more than 17 mm Hg), and none of these patients survived. To predict percent cardiac return preoperatively, the conventional indices of systemic ventricular volume, pulmonary artery area index, arterial blood oxygen saturation, pulmonary blood flow index, and pulmonary vascular resistance were analyzed. None of these showed significant correlation with percent cardiac return. However, all the patients who had a high percent cardiac return (more than 30%) also had both high arterial blood oxygen saturation (more than 75% in room air) and small pulmonary artery area index (less than 55%). In addition, the age at operation showed good correlation ( r = 0.6, p < 0.01) to percent cardiac return. We conclude that high percent cardiac return is one of the risk factors for Fontan operations and can be predicted preoperatively, even if the collateral arteries are not always visible on angiography. If aortopulmonary collaterals are visible on angiography, it is suggested that their preoperative embolization may decrease percent cardiac return and hence operative risk.